Xref: utzoo comp.binaries.ibm.pc.d:9794 comp.sys.ibm.pc.misc:2443
Path: utzoo!utgpu!cunews!cognos!alzabo!tigris!glee
From: glee@tigris.uucp (Godfrey Lee)
Newsgroups: comp.binaries.ibm.pc.d,comp.sys.ibm.pc.misc
Subject: Re: Is AWK up to this application?
Keywords: awk
Message-ID: <1990Oct7.223228.10387@tigris.uucp>
Date: 7 Oct 90 22:32:28 GMT
References: <1661@wjvax.UUCP>
Reply-To: glee@tigris.UUCP (Godfrey Lee)
Organization: Godfrey Lee
Lines: 69

In article <1661@wjvax.UUCP> mario@wjvax.UUCP (Mario Dona) writes:
>Does anyone know of a way of extracting information from a text file
>which contains variable length fields, and outputting it in a different
>format?

Looks more like fixed length fields below.

>For example, I have a text file which contains names and
>addresses as shown below.
>
>John Doe           1563 Meadow Lane         San Jose, CA 94325
>John Smith         345 N. First Street      Oakland, CA 94356
>Henry & Martha Coe 4567 Poplar Ave.         Newark, NJ 78389
>
>^                  ^                        ^
>|                  |                        |_____  column 45
>|                  |______________________________  column 20
>|_________________________________________________  column 1
>
>I want to extract the names and addresses and output it in the following
>format:
>
>John Doe
>1563 Meadow Lane
>San Jose, CA 94325
>
>John Smith
>245 N. First Street
>Oakland, CA 94356
>
>Henry & Martha Coe
>4567 Poplar Ave.
>Newark, NJ 7838
>
>Can this be done using AWK, and if so how? Or is there some other way?


Simple, for example:

	awk -f address.awk <infile >outfile

-------------- address.awk ----------------------------
{
	name = substr ( $0,  1, 19 )
	add1 = substr ( $0, 20, 25 )
	add2 = substr ( $0, 45 )
	printf ( "%s\n%s\n%s\n\n", name, add1, add2 )
}
---------end of address.awk ----------------------------


Of course, it would be simpler if you just separate the fields with a single
character such as ':', e.g.

John Doe:1563 Meadow Lane:San Jose, CA 94325:more stuff --->
John Smith:345 N. First Street:Oakland, CA 94356:more stuff --->
Henry & Martha Coe:4567 Poplar Ave.:Newark, NJ 78389:more stuff--->

then all you need is:

	awk -F: '{ printf ("%s\n%s\n%s\n\n", $1, $2, $3 ) }' <infile >outfile
	    ^^^
	      |
	      +---- defines field separator to be ':'

Good luck.
-- 
Godfrey Lee
cognos!alzabo!tigris!glee